Effects of Schizochytrium and micro-minerals on immune, antioxidant, inflammatory and lipid-metabolism status of Micropterus salmoides fed high- and low-fishmeal diets

A 12-week factorial experiment was conducted to investigate the interactive effects of dietary algal meal (Schizochytrium sp., AM) and micro-minerals (MM, either organic [OM] or inorganic [IM]) on the immune and antioxidant status, and the expression of hepatic genes involved in the regulation of antioxidants, inflammatory cytokines, lipid metabolism, and organ growth of largemouth bass (LMB; Micropterus salmoides) fed high-and low-fishmeal (FM) diets. For this purpose, two sets of six iso-nitrogenous (42% crude protein) and iso-lipidic (12% lipid) diets, such as high (35%) and low (10%) FM diets were formulated. Within each FM level, AM was used to replace 50% or 100% of fish oil (FO), or without AM (FO control) and supplemented with either OM or IM (Fe, Zn, Mn, Cu, and Se). Diets were fed to juvenile LMB (initial weight, 25.87 ± 0.08 g) to near satiation twice daily. The results indicated that FO replacement by dietary AM did not change the levels of most biochemical (ALB, AMY, TP and GLOB), antioxidants (SOD, GPx and GSH), and immune (IgM and lysozyme) parameters in LMB, except ALP and CAT. MM affected only hepatic GSH, with lower values in fish fed the OM diets. FM influenced the levels of ALP, AMY, GLOB, IgM, and MDA (P < 0.05). A three-way interactive effect (P = 0.016) was found on IgM only, with lower levels in fish fed diet 12 (low-FM, AM100, OM). Subsequently, the relative expressions of hepatic antioxidants (Cu/Zn-SOD and GPx-4), inflammatory cytokines (TNF-α and TGF-β1), lipid metabolism (FASN and CYP7A1), and organ growth (IGF-I) related genes were affected by the dietary treatments, with interactions being present in Cu/Zn-SOD, TNF-α, TGF-β1, FASN and IGF-I. Overall, dietary AM could be used as an alternative to FO in low-FM diets without compromising the health of LMB, especially when it is supplemented with MM.


Materials and methods
Diet formulation and preparation. Two sets of six practical diets were formulated to each contain 42% crude protein (CP), 12% lipid, and an estimated 14.64 kJ/g digestible energy (DE). The first set of diets contained 35% FM as the primary protein source (these are considered the high-FM diets) while the second set of diets contained 10% FM (these are considered the low-FM diets). Within each FM level, dietary Schizochytrium sp. meal (AM; Table 1) replaced 50% (AM50) or 100% (AM100) of dietary FO or without AM (AM0, used as FO control), in diets supplemented with Fe, Zn, Mn, Cu, and Se, either in OM or IM forms ( Table 2).
Diet preparation was carried out as follows: solid ingredients were mixed for 30 min using a Hobart mixer (A200; Hobart, Troy, Ohio, USA) followed by the addition of FO and water until the appropriate consistency for pelleting was achieved. The resulting moist diets were passed through a pelletizer with a 3.2 mm die to form pellet strands and air-dried for 24 hours to a moisture content <10%. Diet strands were broken up and sieved using a standard testing sieve (1-mm opening mesh; Fisher Scientific, Pittsburg, PA, USA), sealed in vacuum-sealed plastic bags, and stored in a freezer (−20 °C) until used. A subsample of each diet was collected for proximate and test-mineral composition analyses (Table 2).
Fish, experimental facility and fish rearing. The use of experimental fish was under scientific research protocols of Kentucky State University (KSU) and complied with all relevant local and/or international animal welfare laws, guidelines and policies 30 . In addition, this study was funded by the Alltech-Kentucky State University Alliance (Project No. 17-AAKS-11509), and all the experimental protocols were approved by the Alliance. Juvenile LMB were obtained from F&L Anderson Fish Farm, Lonoke, AR, USA, and transported to the aquatic animal nutrition laboratory (AANL), School of Aquaculture and Aquatic Sciences, KSU, Frankfort, KY, USA; and were acclimatized to the experimental conditions for three weeks. During the acclimatization period, fish were hand-fed to apparent satiation with a 45% CP and 15% crude fat commercial feed (EXTR 450; Rangen Inc., Buhl, ID, USA).
The experiment was conducted in a closed recirculating freshwater aquaculture system (RAS) at AANL. At the commencement of the feeding trial, fish were hand-graded and stocked in each of thirty-six 110-L glass tanks at a stocking density of 15 fish (mean individual weight = 25.87 ± 0.08 g)/tank. Three tanks were randomly assigned to each experimental diet. Fish in each tank were hand-fed with their respective diet to apparent satiation twice daily (at 08:00 and 16:00 hours) for 12 weeks. All water quality parameters were maintained within the acceptable ranges for LMB, including temperature (25.27 ± 0.5 °C), pH (7.56 ± 0.03), dissolved oxygen (6.19 ± 0.04 mg/l), total ammonia nitrogen (0.44 ± 0.02 mg/l), nitrite nitrogen (0.28 ± 0.01 mg/l) and salinity (1.61 ± 0.03ppt). A 12-h photoperiod was provided by artificial lighting controlled by a timer.
Sample collection. At the end of the 12-week feeding trial, sampling was conducted after a 24-h fasting.
Three representative fish from each tank (9 fish per treatment) were anaesthetized with 100 mg/l of tricaine methanesulfonate (MS-222). Blood samples were obtained from the caudal vein using heparinized syringes and transferred to 2.0 ml tubes. Blood samples were centrifuged at 3,000 × g at 4 °C for 10 min. The supernatant was removed and stored at −80 °C for plasma biochemistry and immune parameters assays. Liver samples of the RNA isolation, reverse transcription and real-time quantitative PCR assay. RNA isolation, reverse transcription and real-time quantitative PCR analysis were conducted according to previous studies 9,33 . Briefly, total RNA was isolated from LMB liver samples using a TRIzol ® reagent (Invitrogen ™ , Carlsbad, CA, USA). To avoid genomic contamination, the extracted RNA was treated with RNase-Free DNase (Takara, Dalian, China). The quality and quantity of the isolated RNA were assessed using a NanoDrop ™ One C spectrophotometer (Thermo Scientific ™ , Madison, WI, USA). Complementary DNA (cDNA) was synthesized using the PrimeScript TM RT reagent kit, following the manufacturer's instructions (Takara, Dalian, China). Specific primers for most target genes (growth, metabolism, and immune-cytokine-and antioxidant-related genes) were designed using online resources according to the partial cDNA sequences of the target genes for Micropterus salmoides transcriptome analysis, using the published sequences of LMB or adopted from published articles ( Table 3). All primers for the target genes and housekeeping gene were synthesized by a commercial company (Life Technologies Corporation, Grand Island, NY, USA). Real-time qPCR was used to determine mRNA levels for the target genes and performed according to standard protocols of the manufacturers 33 . Briefly, cDNA (2.0 µL) was reacted with 10.0 µL SYBR Premix Ex Taq II (2X), 0.8 µL forward primer (10 µM), 0.8 µL reverse primer (10 µM), 0.4 µL ROX TM reference dye or dye II (50X), and 6.0 µL RNase-free distilled water in a final reaction volume of 20 µL. The real-time PCR was carried out in a StepOnePlus TM Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The thermocycling conditions for the target genes were initiated with the denaturation step at 95 °C for 30 s followed by forty cycles of 95 °C for 5 s, 60 °C for 34 s, and 95 °C for the 30 s, 95 °C for 3 s, and 60 °C for 30 s. A melting curve analysis was performed (over a range of 50-95 °C) to verify that a single PCR product was produced. Threshold cycle number (C T ) for each sample was determined using Applied Biosystems software and was related to the concentration of the target genes. The housekeeping gene for M. salmoides 23 (β-actin) was used to normalize the expression levels  www.nature.com/scientificreports www.nature.com/scientificreports/ of the target genes. After verifying that the primers were amplified with 100% efficiency, the expression results were analyzed using the 2 −ΔΔCt method 34 .

Statistical analysis. All data were validated for normality and homogeneity of variances by Shapiro-Wilk
and Levene's tests, respectively. A three-way analysis of variance (ANOVA) was performed to detect the interactive effects of dietary FM (high and low), AM (control, and 50% and 100% FO replacements), and MM source (OM and IM). Statistical significance was considered as P < 0.05. If significant interactions were identified, the Tukey's honest significant difference (HSD) test was used for mean separation. In the absence of significant interactions (additive model), the means of any significant factor with more than two levels were separated using Tukey HSD test. All statistical analyses were performed using Statistical Analysis System software (SAS; SAS Institute Inc., Cary, NC, USA). Results are expressed as mean ± standard error (SE).

Results
Plasma biochemical and immune parameters. Plasma biochemical (ALP, AMY, GLOB, ALB and TP) and immune (IgM and lysozyme) parameters are presented in Table 4. LMB fed low-FM diets displayed significantly higher ALP activity (P = 0.001), while the reverse was found for AMY activity (P = 0.009) and GLOB concentration (P = 0.020). Replacement of dietary FO by AM influenced the activity of ALP only, with its lower level in fish fed AM100 (P < 0.05), but not significantly different from the FO control (AM0). LMB fed diets 5 (high-FM, AM100, IM) and 6 (high-FM, AM100, OM) had lower ALP activities compared to fish fed diet 8 (low-FM, OM) while fish fed diet 6 also had lower plasma ALP activity than those fed diet 3 (high-FM, AM50, IM) (P = 0.004). A two-way interaction between dietary FM and MM was observed for ALP activity and GLOB concentration. There were no three-way interactive effects of FM, AM, and MM on the plasma biochemical parameters.
A three-way interaction (P = 0.016) of FM, AM and MM was detected for IgM, with its higher concentration in the plasma of LMB fed diet 6 (high-FM, AM100, OM) compared to those fed diet 12 (low-FM, AM100, OM), while no significant differences were found among other treatments. Plasma lysozyme was not significantly affected by dietary treatments.
Hepatic peroxide and antioxidants. Hepatic peroxide and antioxidant parameters including MDA, SOD, GPx and GSH, were not significantly affected by replacement of dietary FO with AM and no interactions were found (Table 5). Significantly higher MDA content was observed in the liver of LMB fed high-FM diets compared to those fed low-FM diets. LMB fed AM0 diets displayed lower (P = 0.002) CAT activity relative to those fed AM50 and AM100 diets. A slightly but significant effect of dietary MM was observed for GSH activity in the liver of LMB, which was higher in groups fed IM than those fed OM (P = 0.042).

Relative expression of hepatic antioxidants genes. The relative expressions of antioxidants genes
in LMB fed the experimental diets are shown in Figs. 1 and 2. A three-way interactive effect (P = 0.030) of dietary FM, AM, and MM was found on the relative gene expression of Cu/Zn-SOD, which was significantly down-regulated in fish fed diet 2 (high-FM, AM50, OM) compared to those fed diet 8 (low-FM, AM0, OM) (Fig. 1). The gene expression of hepatic glutathione peroxidase 4 (GPx-4) was affected (P < 0.001) by dietary AM only, with the lowest level observed in LMB fed AM100 (Fig. 2).  www.nature.com/scientificreports www.nature.com/scientificreports/ Relative expression of hepatic pro-and anti-inflammatory cytokines genes. Figures 3 and 4 presented the relative expressions of hepatic pro-and anti-inflammatory cytokines genes in LMB fed the experimental diets. Two-way interactive effects of dietary FM and AM were obtained on the relative gene expression of hepatic tumor necrosis factor alpha (TNF-α) (P < 0.001) and transforming growth factor-β1 (TGF-β1) (P = 0.011). In contrast to the other treatments, fish fed high-FM and AM0 diets had significantly lower TNF-α expression except compared those fed low-FM and AM50 diets (Fig. 3). Low-FM with AM0 level down-regulated the TGF-β1 expression compared to high-FM (with AM50, and AM100) and low-FM with AM100 treatments (Fig. 4).
Relative expression of genes related to hepatic lipid metabolism and organ-growth. The relative expressions of genes related to hepatic lipid metabolism are shown in Figs. 5 and 6. Interactive effects (P = 0.017) of dietary FM and AM were detected on the relative gene expression of hepatic fatty acid synthase (FASN) in LMB fed the experimental diets (Fig. 5). Dietary AM up-regulated the expression of FASN in fish fed the low-FM supplemented with AM50 and AM100 diets relative to those in low-FM with AM0 groups. AM was the only factor that significantly affected the expression of cholesterol 7-alpha-monooxygenase (CYP7A1) in LMB, which was significantly up-regulated in the groups fed AM100 in contrast to those fed AM0 and AM50 (Fig. 6). Figure 7 presented the relative expression of insulin-like growth factor I (IGF-I), a gene related to hepatic-growth in LMB fed the experimental diets. In general, LMB fed high-FM diets displayed greater expression of IGF-I than those fed low-FM diets (P = 0.007), while significantly higher expression of IGF-I was also found in LMB fed the low-FM supplemented with AM50 diets relative to low-FM with AM100 fed groups. These responses were not consistent across all treatments and a three-way interaction was observed (P = 0.022). The  www.nature.com/scientificreports www.nature.com/scientificreports/ relative expression of IGF-I was significantly up-regulated in the group fed diet 2 (high-FM, AM0, OM) compared to those fed diets 6 (high-FM, AM100, OM), 7 (low-FM, AM0, IM), 8 (low-FM, AM0, OM), 11 (low-FM, AM100, IM), and 12 (low-FM, AM100, OM), while not significantly different from the remaining treatments (Fig. 7).  Table 5. Hepatic peroxide and antioxidants of LMB fed the experimental diets for 12 weeks a . MDA, malondialdehyde; SOD, superoxide dismutase; CAT, catalase; GPx, glutathione peroxidase; GSH, glutathione. a Data are mean value ± SE. We performed nine analysis per treatment; n = 9. Mean values in the same column with different superscript are significantly different (P < 0.05).

Discussion
In aquatic animals, blood parameters are considered as important indicators for physiological conditions and health status in response to dietary treatments 8,9,35 . In this study, we did not find any significant three-way interactive effects of FM, AM, and MM on any analyzed plasma biochemistry parameters in LMB. Simultaneously, the replacement of dietary FO by AM did not influence most of the plasma biochemistry in LMB, except ALP. Similarly, no significant effects of dietary Schizochytrium sp., supplemented with either inorganic or organic MM (OM: Zn, Cu, Mn, Fe, and Se) were observed on blood plasma chemistry of Atlantic salmon 7 . Our recently published study revealed that dietary Schizochytrium meal could replace dietary FO up to 75% and play an imperative role as a source of essential fatty acids in shrimp (Litopenaeus vannamei) diets without compromising growth and health 4 . However, in the present study, there were significant differences in blood plasma ALP levels between treatments with 50% and 100% dietary FO replacements by AM, with ALP levels higher in fish fed AM50, but these were not significantly different compared to the AM0 treatments. Nonetheless, the highest level of plasma ALP in our study (87.0 U/L) is below the normal ranges that reported in Atlantic salmon (647-988 U/L) 36 . In humans, the level of ALP in a healthy adult range from 20-140 U/L; however, children tend to have significantly   www.nature.com/scientificreports www.nature.com/scientificreports/ higher levels of ALP than adults because their bones are still growing 37 . The presence of ALP activity in plasma is directly related to the release of ALP enzymes from cells to the extracellular fluids and elevated ALP activity may occur when there is cell growth, tissue necrosis, or leakage of preformed ALP 38,39 . Therefore, in this study, the higher levels of plasma ALP in LMB fed AM50 could be related to the cell/ bone growth of juvenile LMB. Nevertheless, further study is needed to understand the specific mechanisms of the effects of dietary AM on plasma ALP in fish.
In teleost fish, producing immunoglobulins is a specific immune response after being stimulated by antigen and the IgM class is the predominant immunoglobulin in most fish species 40 . Lysozyme is a ubiquitous bacteriolytic enzyme that is part of the nonspecific defense mechanisms in most animals 41 . In this study, FO replacement by dietary AM did not affect the levels of the immune parameters such as IgM and lysozyme, in the plasma of LMB. However, LMB fed diet 6 (high-FM, AM100, OM) displayed significantly higher plasma IgM in contrast to those fed diet 12 (low-FM, AM100, OM), but not among other treatments. We also found interactive effects of dietary FM, AM and MM on the IgM content in LMB. Our findings suggest that dietary AM could maintain immune defense system/ health of LMB, even when replacing 100% of dietary FO, especially when supplemented with MM. This is in agreement with a study on Atlantic salmon that reported whole Schizochytrium sp.-based   www.nature.com/scientificreports www.nature.com/scientificreports/ microalgae can provide a good alternative to the depleting marine resources of n-3 LC-PUFA without detrimental health effects 7,42 , even more pronounced for salmon fed OM 7 . Nevertheless, this is the first study to investigate the interactive effects of three factors (FM, AM and MM) in fish, therefore further study is needed.
Fish hepatic tissue has large quantities of PUFAs 11 , which implies a high risk of oxidative stress since these lipids are major targets for ROS 12,13 . The antioxidant defense system helps fish to maintain endogenous ROS at relatively low levels and to attenuate the oxidative damage induced by the high reactivity of ROS 14 . An increase in free radicals causes overproduction of MDA, which is one of the final products of lipid peroxidation in the cells. Thus, the MDA level is commonly known as a marker of oxidative stress 43 . Nogueira et al. 44 reported that a decrease in MDA content is associated with an increase in antioxidant enzymes in the defense system, which indicates MDA is a toxic substance that is caused by free radicals. The key antioxidants in the antioxidant defense system include SOD, GPx, CAT, and GSH 13 . In fish, antioxidant enzymes activities correlate with nutritional factors 9,45 . In the present study, we did not find significant differences between the control (FO control) and AM treatments in the hepatic MDA, SOD, GPx and GSH levels of LMB, except higher CAT activity in LMB fed AM-diets. The significantly higher MDA content observed in the liver of LMB fed high-FM compared to low-FM diets might be related to the MDA content of dietary FM, which was not quantified in this study. Whereas IM supplementation supported significantly higher GSH in comparison to OM (1.37 ± 0.01 vs 1.35 ± 0.00), such narrow differences suggest limited practical significance.
The activities of antioxidant enzymes in fish are closely related to their corresponding gene expression levels 46 . In this study, there were no significant differences in the level of hepatic Cu/Zn-SOD mRNA among treatments, except the down-regulation in the high-FM treatment (with AM0 and OM). A three-way interactive effect (P = 0.030) of dietary FM, AM, and MM was found on Cu/Zn-SOD mRNA level. The gene expression of hepatic GPx-4 was significantly affected by dietary AM only, with its lowest level in LMB fed AM100 diets. The antioxidants mRNA levels showed similar trends to the activities of antioxidant enzymes in this study. The regulation of hepatic antioxidant genes expression in LMB fed various dietary treatments in this study could be related to the nuclear factor erythroid 2-related factor 2 (Nrf2) signaling pathway that play a vital role in the regulation of antioxidant enzymes gene transcriptions in fish 9 . Furthermore, though we did not find significant effect of MM source as a single factor (organic and inorganic) on the antioxidant defense of LMB, the interactive effects of FM, AM and MM on the Cu/Zn-SOD mRNA level in this study suggest that the MM (individually or in combination) could be involved in the antioxidant defense of LMB fed low-FM diets. Other studies reported that dietary MM supplemented in OM forms were more effective than IM forms in the antioxidant defense of other fish species 5,47 . The possible reason for the different findings of these studies could be due to differences in experimental design, fish species, minerals source and other factors.
TNF-α is pro-inflammatory cytokines, whereas TGF-β1 is anti-inflammatory 9,48 . In this study, interactive effects of dietary FM and AM were found on the relative gene expression of hepatic TNF-α and TGF-β1. Concurrently, an inverse trend was observed between the relative gene expressions of TNF-α and TGF-β1, with dietary AM (especially at 50% FO replacement) down-regulating TNF-α mRNA and up-regulating TGF-β1 in fish fed low-FM diets; although not significant for the former. The regulation of these genes by dietary FM and AM could be mediated by the signaling molecules of nuclear transcription factor-κB (NF-κB) 49 and target of rapamycin (TOR) 8 , although this needs further elucidation.
FASN is a key enzyme for the de novo synthesis of fatty acids 50 . CYP7A1 is involved in cholesterol metabolism or elimination 51 through catalyzing the first and rate-limiting step in the classic pathway of bile acid synthesis 52 . In this study, AM up-regulated the mRNA levels of hepatic FASN in LMB fed low-FM diets. Interactive effects of dietary FM and AM were also detected on the relative gene expression of hepatic FASN, while the expression of CYP7A1 mRNA was regulated by dietary AM only, with its augmented levels in LMB fed AM100 diets. A previous study reported that the mechanism of bile acids regulating the CYP7A1 expression is mainly through the farnesoid X receptor signaling (FXR)/ small heterodimer partner (SHP) pathway (FXR/SHP pathway) in the liver 53 . Our study proved that dietary AM could play an important role in lipid metabolism and cholesterol elimination in the liver of LMB through de novo synthesis of fatty acids and bile acids. Nevertheless, to our knowledge, this is the first report regarding the effect of dietary AM on FASN and CYP7A1 mRNA levels in the hepatic cells of fish, and the specific mechanisms behind the regulation of de novo synthesis of fatty acids and bile acids (FASN and CYP7A1 augmentation), cholesterol elimination, and promotion of dietary lipids metabolism in LMB needs further investigation.
Furthermore, in previous studies, the relative expression levels of the GH-IGF system were effectively used in the evaluation of organ growth and, ultimately, fish growth in response to different dietary treatments and alternative feed ingredients 33,54,55 . In this study, the relative expression of hepatic IGF-I was regulated by the dietary treatments. Specifically, IGF-I was significantly augmented in LMB fed the high-FM diet (with AM0 and OM) but was not different from those fed the low FM and AM50 diets. Interestingly, interactive effects of FM, AM, and MM were found on the relative expression of IGF-I mRNA. Thus, the relative expression of this signaling molecule could supply some important insights in understanding the underlying mechanisms for liver growth and, ultimately, the growth of LMB fed low FM diets supplemented with AM and MM.

Conclusion
The results of this study indicated that dietary AM could completely replace FO (even in low-FM diets) and improve/ maintain the immune, antioxidant, anti-inflammatory, and lipid metabolism capacity of LMB. Similarly, the MM (individually or in combination) could be involved in the antioxidant defense of LMB fed low-FM diets. Indeed, dietary AM can be used as a good alternative to FO without detrimental impacts on the health of LMB, especially when it is supplemented with MM. Moreover, the relative expressions of hepatic Cu/Zn-SOD, GPx-4, TNF-α, TGF-β1, FASN, CYP7A1, and IGF-I mRNA were regulated by the dietary replacements, with interactive effects on the mRNA levels of Cu/Zn-SOD, TNF-α, TGF-β1, FASN, and IGF-I. These signaling molecules could